We propose to establish a series of methods suitable to recognize the development of pneumotoxicity in laboratory animals, rats and guinea pigs. These methods will rely on measurement of respiratory patterns of animals during ambient air breathing and during CO2 challenges. With the first method, we will be able to recognize that pulmonary toxicity is developing by repeated measurements in the same animals. This method will use a whole body plethysmograph which will permit measurement in unrestrained, unanesthetized animals in a comfortable environment. Following detection of pulmonary toxicity, the animal is then tested using other measurements to establish the type of toxicity induced. The methods proposed can be used to detect toxicity development for both airborne or systemically administered chemicals and will detect acute as well as chronic damage. The chemicals involved in evaluating how effective the methods we propose to use will include the following: cotton dust, silica, paraquat, bischloroethylnitrosourea, hexamethylene diisocyanate trimer, histamine, serotonin, carbamylcholine, propranolol and sulfuric acid mist. We propose to correlate the results obtained with the data in humans available from the literature and establish models which will have predictive power capabilities. Of particular importance, the methods proposed are adaptable in toxicological studies where it is desirable to make repeated measurements in an economical fashion. Also the number of animals required to obtain concentration-response relationships is very low in comparison to existing methods.